# -*- coding: utf-8 -*-

"""
Python3.0

Created on Sat Nov 26 08:54:26 2016

需要的安装包

pip install pytagcloud

pip install pygame

pip install simplejson

@author: daxiong

"""

import pytagcloud

#wordcounts 是一个列表，元素是元组

wordcounts=[("python",5),("love",2),("work",1)]

tags = pytagcloud.make_tags(wordcounts)

pytagcloud.createtag_image(tags, 'cloud_large.png', size=(900, 600))

# -*- coding: utf-8 -*-

"""

Created on Mon Aug 14 15:55:43 2017

@author: toby

"""

from wordcloud import WordCloud

import matplotlib.pyplot as plt

text = '''文案 文案

The  抱抱 Zen of LOVE 抱抱 Python, 快乐 by Tim Peters

公众号 公众号 Python 最好的 语言 语言

一辈子 is better LOVE than 一辈子.

喵小姐 is 爱你 than  implicit.爱你 喵小姐

蟹先生 is 爱你 than complex.

一辈子 is 蟹先生  than complicated.

二中 is 喵小姐 我想你了 than nested. 二中 蟹先生

清湖 is 胜于 than 清湖.

思旺 counts. 想你

Special 喵小姐 我想你了 aren't special enough 思旺 break 思旺 rules.

别生气 practicality beats 厨艺好.

Errors should 我想你了 never pass 小龙虾 silently. 运营

别生气 explicitly 好不好. LOVE

In the face of ambiguity, 程序员 the 厨艺好 to guess.龙华 龙华

There 快乐 should be one-- 我想你了 and preferably 红烧肉 only one 小龙虾--obvious way to do it.运营

Although 共享单车 way may not 我想你了 be obvious at first unless you're Dutch. 新媒体 地铁

Now is better 红烧肉 than never.

程序员 Although 共享单车 is often 高铁 than 东莞 now. 高铁 地铁

If the implementation 想你 is hard to explain, it's a bad idea. 想你了

If 成都 implementation is 想你 easy to explain, it may be a good idea.

Namespaces are 端午one 端午 honking great idea -- 成都 do more of those! 想你了

深圳 晚安 深圳 新媒体

'''

# the font from github: https://github.com/adobe-fonts

font = r'C:\Windows\Fonts\simfang.ttf'

wc = WordCloud(collocations=False, font_path=font, width=1400, height=1400, margin=2).generate(text.lower())

plt.imshow(wc)

plt.axis("off")

plt.show()

wc.to_file('show_Chinese.png')  # 把词云保存下来

# -*- coding: utf-8 -*-

"""

Created on Mon Aug 14 08:57:21 2017

@author: toby

"""

import jieba

import pandas

import urllib

import re

from bs4 import BeautifulSoup

from wordcloud import WordCloud

import matplotlib.pyplot as plt

#引用微软的中文字体

font = r'C:\Windows\Fonts\simfang.ttf'

'''

#赛柏蓝

url="http://mp.weixin.qq.com/profile?src=3&timestamp=1502674891&ver=1&signature=wDxrEoM1f5I3Js7rVZL7XeAOVS5q6FoHLVuMdM*iJLccLjB80A1BESmnMfk62BrS9Uz4VyJ05uzI8aJoW6r6Vw=="

#医药魔方

url="http://mp.weixin.qq.com/profile?src=3&timestamp=1502690954&ver=1&signature=*jfLnFfVyfWmIcIyMN*R4*av27A-ubJUNLoiD2B17*Zqj9W*HPoPKdtFegntGr0Ft7jVfdMcW9tLMjBXf7r4vQ=="

'''

url="http://mp.weixin.qq.com/profile?src=3&timestamp=1502691348&ver=1&signature=piBaU5ZN*TbBiF41yhw-D4sDK9*mY8TOht4snXuo6Hrm3UKMe5I7wpU*zgCa2Bk3DS-joq3oSNOXboucANMwWg=="

html = urllib.request.urlopen(url).read()

text = BeautifulSoup(html,"lxml").get_text()

#正则表达式规则

#标题名

regex = re.compile(r'\"title\"\:\"(.+?)\"')

#摘要

regex1 = re.compile(r'\"digest\"\:\"(.+?)\"')

#print(regex.findall(text))

#print(regex1.findall(text))

len1=len(regex.findall(text))

list_tilte=regex.findall(text)

list_digest=regex1.findall(text)

#把标题和摘要合并在一起

list_allWords=list_tilte+list_digest

filename="nonesense_words.txt"

#list_noneSense_words=["赛柏蓝","出品","我们","一个","工作","问题","中国","大家"]

#垃圾词库

def List_none_sense(filename):

    file_noneSense_words=open(filename,'r')

    list_noneSense_words=[]

    for line in file_noneSense_words:

        line_list=line.split()

        for word in line_list:

            list_noneSense_words.append(word)

    return list_noneSense_words

#判断一个单词是否在垃圾词表里

def none_sense_words_Judeg(word):

    if word in list_noneSense_words:

        return True

    else:

        return False

#过滤停用词列表

def filter_none_sense_words(list1):

    list2=[]

    for i in list1:

        #如果不在垃圾词库里或不是数字,汉字长度大于1

        if none_sense_words_Judeg(i[0])==False and i[0].isdigit()==False and len(i[0])>1:

            #print("remove",i)

            list2.append(i)

    return list2

def Fenci():

    list_total=[]

    for i in list_allWords:

        seg_list = list(jieba.cut(i,cut_all=False))

        list_total+=seg_list

        #生成一个字典，每个单词排名

        tf={}

        for seg in list_total :

            #print seg

            seg = ''.join(seg.split())

            if (seg != '' and seg != "\n" and seg != "\n\n") :

                if seg in tf :

                    tf[seg] += 1

                else :

                    tf[seg] = 1

    return sorted(tf.items(),key=lambda item:item[1],reverse=True)

#筛选排名大于三的，并且名词长度大于1

def more3_filter(list1):

    list2=[]

    for i in list1:

        if i[1]>2 and len(i[0])>1:

            list2.append(i)

    return list2    

#垃圾词表

list_noneSense_words=List_none_sense(filename)

list_words=Fenci()

#过滤停用词

list_words1=filter_none_sense_words(list_words)

list_words2=more3_filter(list_words1)

#前二十

list_words3=list_words2[:20]

for i in list_words3:

    print (i)

#写入数据到Excel，用pandas的df数据结构

df=pandas.DataFrame(list_words3)

df.to_excel("result.xlsx")    

#标签云传输的frequency必须是字典形式，所以要转换

list_words3=dict(list_words3)

wc = WordCloud(collocations=False, font_path=font, width=1400, height=1400, margin=2).generate_from_frequencies(list_words3)

plt.imshow(wc)

plt.axis("off")

plt.show()

wc.to_file('show_Chinese.png')  # 把词云保存下来